Interesting Links for 05-08-2017

[andrewducker]

A Dinosaur So Well Preserved It Looks Like a Statue

(tags: dinosaurs archeology )

Have Smartphones Changed The Latest Generation? - The Atlantic

(tags: phones Technology society children )

The Ctenophore evolved it's brain entirely separately to all other life

(tags: evolution brain life )

It’s not that your teeth are too big: your jaw is too small

(tags: teeth evolution )

Tryptophan may help soothe inflamed gut

(tags: mice microbiome )

Nuclear is the safest major energy source

(tags: nuclearpower safety )

Math Journal Editors Quit to Start Competing Open Access Journal

(tags: academia publishing OpenAccess )

Comments

[drplokta]

On the third link, I think you should have said ctenophore, not carnivore. Auto-correct playing up?

[andrewducker]

Dammit, yes!

Will fix from home!

Smartphones.

[ironyoxide]

Whenever anyone suggests that making it easier for ordinary people to communicate with each other is a bad idea, I get suspicious.

Re: Smartphones.

[andrewducker]

I don't think it's saying that.

Math Journal Editors Quit to Start Competing Open Access Journal

[jack]

I'm not really following the saga of mathematicians jumping ship from paid journals, but I see snippets of if like this. It makes sense that maths is one of the subjects its easiest to escape, since a lot of things get preprints on arxiv anyway, and because there's the least overhead to research. I guess it's significant that this is a journal leaving springer (who have a fairly long history of respected academic publishing) not elsevier (who everyone hates).

Re: Math Journal Editors Quit to Start Competing Open Access Journal

[njj4]

Yes, this happens from time to time. My PhD supervisor (Colin Rourke) helped set up Geometry & Topology twenty years ago along with some of the editors of the long-running, respected journal Topology which was now owned by Elsevier. A couple of years later, most of the editorial board of Topology and its Applications quit to set up Algebraic & Geometric Topology. Both became very successful, and partially eclipsed the journals they'd been set up to compete with. Joan Birman (an eminent topologist who celebrated her 90th birthday a couple of months ago) wrote an interesting article about it at the time.

Initially run on an "open access for individuals but institutions are supposed to pay a modest subscription fee" model, this turned out to be unworkable in practice (otherwise sympathetic librarians found it hard to justify paying an effectively optional subscription for a weird new journal when fees for established journals were going up). So they now run on a "modest subscription fee and everything older than four years is open access" model, which seems to work.

A few years in (round about 2004, I think) this spawned a new not-for-profit academic publishing company, Mathematical Sciences Publishers, which subsequently brought out a few more journals and took over the publication (but not the ownership) of the long-established Pacific Journal of Mathematics. I worked for them for several years (my career in IT had literally just gone up in flames and I was looking for an opportunity to go back into academia, having just finished my PhD), mostly doing LaTeX production editing, Perl programming and miscellaneous system administration for the Warwick-based side of the operation (most of the rest was based in Berkeley, California).

About ten or eleven years ago, the editorial board of Topology resigned en masse, to set up a new journal with the catchy title of The Journal of Topology. Most of the mathematical community then boycotted Topology in favour of G&T and JTop. Elsevier tried to keep it afloat by hunting around for a new editorial board and slowly eking out the papers still in the publication queue, but ultimately they failed and Topology closed in 2009 after nearly fifty years in publication.

I keep thinking that somebody should write a book about all this, along the lines of Ben Goldacre's Bad Pharma, but nobody seems to be doing so yet. I'm slightly tempted to give it a go, after I've finished my current book and the next one on my list.

Re: Math Journal Editors Quit to Start Competing Open Access Journal

[andrewducker]

You could start by writing it as a set of blog posts. Which, if they were anything like the above comment, would make fascinating reading.

[randomdreams]

I feel like the primary issue with nuclear isn't safety per se but that it's so astonishingly expensive to build one, even when over 20 years it'll repay itself, that it's hard to convince anyone that it makes economic sense to do so, and particularly hard to convince anyone to spend the extra money on ongoing safety systems/procedures/investigation.

[randomdreams]

And what's even more frustrating is that many of the designs that would result in vastly lower levels of nuclear waste being produced haven't been proven on a commercial level because A: they're even less economically viable and B: they have a vastly higher risk of being manipulated to produce atomic weapon material rather than burning off their own wastes.

[drplokta]

But it would be a lot cheaper to build if we lowered the required safety standards so that it was only a bit safer than the next-safest alternative. Safety is expensive. It's no coincidence that nuclear power is both very safe and very expensive.

[randomdreams]

While I agree, I also feel that because nuclear problems can be a lot harder to clean up than failed dams or coal waste floods, holding them to a high standard of safety is prudent. For one thing, whether justifiably or not, adjacent property owners are going to see the value of their property annihilated if there's a serious problem.

[danieldwilliam]

Along with safety considerations which add to the cost I think there are also three other things going on. One is the size of the power stations mean they are built in small numbers. They are usually the result of a national nuclear programme and even in large developed nations there is usually only room for a handful of new nuclear plants in each generation. You therefore struggle to gain much in the way of scale economies.

Secondly, there seems to be a lot of tweaking of designs within programmes so that often it appears that your nuclear fleet is actually a series of proto-types with the associated costs of doing things for the first time.

The third factor is that with large, long term projects with huge capital requirements comes a long-term financial risk which makes funding difficult or requires long-dated (and politically difficult) guaranteed off-take arrangements.

Smaller modular designs which you can make in a factory like Liberty Ships would seem to offer a way forward. As would buying double-digit numbers of the same design from the Russians or the Chinese. The former seems more politically acceptable in the West.

That said, I'm still of the opinion that renewables plus storage prove cheaper than nuclear plus storage and on a timescale so quick that the nuclear industry struggle to react.

[randomdreams]

My understanding of the economics of nuclear is that because of the associated infrastructure, you have to make them enormous to recoup the investment in an acceptable timeframe, and nobody needs that many enormous power plants, so it's hard to get to economy-of-scale levels. (Especially when there are a half zillion designs out there.) I was kinda hoping tiny installations like the Toshiba 4S were going to be the way of the future, but it doesn't seem like that's going to happen.

[danieldwilliam]

I think it depends what infrastructure you are thinking off.

The Chinese are looking at a pebble bed modular reactor based on a German design of about 250 MW. One of the applications for it is to replace the power train in coal power plants by running them in parallel.

Lots of people appear to be looking at modular designs but it's difficult attracting financing in a world where gas and solar PV cheap.

The Indian government has recently approved a fleet of 900MW nuclear plants. Which is something you can do if, like India, you have a population of more than billion and a goodly fraction of them without any electricity at all.

[randomdreams]

Specifically, on-site wet and dry storage for used rods, security, and crashing-aircraft-proof containment, was the infrastructure I was thinking about. Gas, coal, wind, and solar really shine on those fronts, and hydro does pretty well.

I keep reading about the Indian thorium reactors. I lived near a reactor capable of using thorium (and actually did so) and have always thought it would be a good idea to pursue that. (Particularly for India.)

[danieldwilliam]

On the other hand there is some suggestion that nuclear power might be about to get cheaper.

A levelised cost of electricity of about $60 / MWH is my ballpark figure for being in the game.

http://www.theenergycollective.com/dan-yurman/2409802/study-finds-advanced-reactors-will-competitive-costs

[armiphlage]

With wind at $24-$50/MWh and solar at $60-114/MWh (depending on whose numbers you believe), with prices dropping continually, nuclear will have to get much cheaper very rapidly to compete. Plus solar and wind can be built incrementally, reducing risk to investors. I agree with you that renewables plus storage is probably going to be more affordable. Utilities will find it hard to issue bonds for nuclear plants with other options available.

[danieldwilliam]

Are those renewable prices round the wrong way?

I've seen solar PV project below $30 / MWH but no wind project that low.

I think it may turn out to be the case that a small amount of modular nuclear on a grid avoids a lot of battery storage. The battery storage that I think it avoids is the stuff that is used least. The storage of summer sun solar PV for use in February.

AT $60 / MWH or more, some nuclear plant might be cheaper than storage. Depends obviously on the cost of the storage technology, how much hydro there is on the grid and how interconnected the grids are.

[danieldwilliam]

The ankylosuarus, or to give it it's full Captain Scientific Name, the Boomawhacka, is one of my favourite dinosaurs.